196  Cotton- Seed  Oil  and  Beej  Fat  in  Lard.  {AmApXiSarm' 
respectively,  and  although  the  samples  responded  clearly  to  the  general 
tests,  on  applying  the  silver  test  in  the  different  forms  prescribed,  no 
reduction  occurred  alone  or  when  mixed  with  genuine  lard.  Here 
then  we  have  a  serious  objection  to  these  silver  tests,  for  the  matter 
causing  reduction  of  silver  nitrate  seems  to  be  destroyed  by  keeping  j 
hence  a  reaction  with  silver  nitrate  proves  the  presence  of  cotton  oil, 
but  no  reaction  .does  not  prove  the  absence  in  lard. 
Chemists  having  old  samples  of  oil  in  their  possession  will  render 
service  by  trying  the  silver  nitrate  reaction  on  the  oils.  It  is  to  be 
remarked  here  that  I  tried  the  reaction  on  the  fatty  acids  produced  by 
saponification,  with  a  negative  result.  By  acting  on  the  oil  by  sul- 
phur chloride,  small  percentages  of  cotton  oil  can  be  detected. 
Researches  on  this  subject  have  been  published  by  Mr.  Warren. 
Some  time  ago  the  author  made  some  experiments  on  the  surface  ten- 
sion tests  for  oils.  Pure  dry  melted  lard,  dropped  on  water  of  100°  F., 
does  not  extend  or  increase  in  size.  Beef  fat  or  mutton  fat  extends 
over  the  surface  of  the  basin,  breaking  up  into  innumerable  droplets, 
and  is  agitated  by  a  rapid  gyratory  motion. 
The  author  mixed  pure  lard  with  10,  15,  20  and  25  per  cent,  of 
beef  fat,  and  allowed  one  drop  to  fall  on  water  of  100°  F.  By 
noticing  the  degree  of  expansion  and  the  time  required,  approximate 
determinations  of  the  beef  fat  can  be  obtained.  Mutton  fat  behaves 
similarly,  and  therefore  cannot  be  distinguished.  On  repeating  this 
with  exactly  the  same  lard  eight  months  after,  the  drop  behaved 
exactly  like  beef  or  mutton  fat,  this  being  most  unfortunate.  The 
only  other  test  is  the  delicate  test  with  iodo-mercuric  chloride  of  HubL 
This  method  is  one  of  the  most  delicate  and  best  in  the  chemistry  of 
oils  and  fats.  The  author's  experience  is  most  extensive,  its  results 
are  exact  and  reliable.  Somewhat  discrepant  statements  exist,  how- 
ever, as  to  the  iodinic  absorption  of  cotton-seed  oil  and  lard.  Thus 
Pattinson  {Journal  Society  Chem.  Industry,  No.  1,  vol.  viii.)  gives 
from  57  to  63  per  cent,  iodine  absorbed  by  lard,  and  105  to  116  per 
cent,  for  cotton  oil.  The  author  has  never  found  more  than  60  per 
cent,  for  lard  and  110  per  cent,  for  cotton-seed  oil.  Again,  Pattinson 
gives  41  per  cent,  for  beef  fat,  whilst  my  figures  approach  the 
average,  44  per  cent.  It  is  needless  to  say  these  different  statements 
would  cause  grave  errors  in  calculating  the  foreign  fat  in  a  sample  of 
lard.  It  would  be  of  great  service  if  chemists  would  state  their 
experience  of  this  most  excellent  test.    In  my  opinion  no  satisfactory 
